Real-Time, Highly Accurate Robotic Grasp Detection using Fully Convolutional Neural Networks with High-Resolution Images

TL;DR

This paper introduces a fully convolutional neural network approach for robotic grasp detection that achieves high accuracy and real-time performance on high-resolution images, enabling effective grasping of novel objects.

Contribution

The paper presents a novel FCNN-based method for robotic grasp detection that is both highly accurate and capable of real-time processing on high-resolution images, adaptable to various image sizes.

Findings

Achieved 96.6% detection accuracy on Cornell dataset.

Real-time processing at 6-20ms per image.

90% success rate in grasping challenging small objects.

Abstract

Robotic grasp detection for novel objects is a challenging task, but for the last few years, deep learning based approaches have achieved remarkable performance improvements, up to 96.1% accuracy, with RGB-D data. In this paper, we propose fully convolutional neural network (FCNN) based methods for robotic grasp detection. Our methods also achieved state-of-the-art detection accuracy (up to 96.6%) with state-of- the-art real-time computation time for high-resolution images (6-20ms per 360x360 image) on Cornell dataset. Due to FCNN, our proposed method can be applied to images with any size for detecting multigrasps on multiobjects. Proposed methods were evaluated using 4-axis robot arm with small parallel gripper and RGB-D camera for grasping challenging small, novel objects. With accurate vision-robot coordinate calibration through our proposed learning-based, fully automatic approach, our proposed method yielded 90% success rate.